Integrated self-driven active clamp

- Appulse Power Inc.

A method involves receiving, at an active clamp controller circuit, an active clamp switch current that passes through an active clamp switch. The active clamp switch is enabled using the active clamp controller circuit in response to determining, based on the active clamp switch current, body-diode conduction of the active clamp switch. The active clamp switch is disabled using the active clamp controller circuit in response to determining, based on the active clamp switch current, a first zero-crossing of the active clamp switch current and a second zero-crossing of the active clamp switch current.

Skip to: Description  ·  Claims  ·  References Cited  · Patent History  ·  Patent History
Description
RELATED APPLICATIONS

This application is a continuation of U.S. patent application Ser. No. 16/913,489, filed on Jun. 26, 2020, which issued as U.S. Pat. No. 11,095,228 on Aug. 17, 2021, which is a continuation of U.S. patent application Ser. No. 16/145,819, filed on Sep. 28, 2018, which issued as U.S. Pat. No. 10,707,766 on Jul. 7, 2020, and which claims priority to U.S. Provisional Patent Application No. 62/625,691, filed on Feb. 2, 2018, and entitled “Integrated Adaptive Active Clamp,” all of which are hereby incorporated by reference for all purposes.

BACKGROUND

Switch-mode power supplies (SMPS) are power management components in modern electronic devices. They provide, among other things, power efficient and galvanically isolated power to multiple loads. To achieve high power processing efficiency and/or galvanic isolation, conventionally one or more magnetically coupled elements, semiconductor switches and associated gate driver circuits are required.

The magnetically coupled elements often suffer from non-trivial leakage inductance phenomena, which necessitate the need for affordable voltage snubber circuits to control the semiconductor switch peak drain-to-source voltages. Because of the price-sensitive nature of SMPS, the snubber circuits are conventionally limited to the cost-effective passive and power lossy resistor-capacitor-diode (RCD) configurations.

In systems sensitive to power losses and heat generation, the dissipation in lossy components in the form of heat is unsuitable. Thus, recycling of energy using an active clamping configuration within the system provides an opportunity for system form-factor reduction and power efficiency improvement.

Additionally, clamping the maximum drain-source voltages of switching power transistors allows for increased device reliability and use of power transistors with improved figure-of-merit (FOM). The improved FOM enables the SMPS to operate at higher switching frequency while maintaining high power processing efficiency. Furthermore, it allows for a reduction of the SMPS reactive component size and cost.

SUMMARY

In some embodiments, an active clamp circuit includes an active clamp capacitor coupled in series with an active clamp switch and an active clamp controller circuit. The active clamp controller circuit receives an active clamp switch current that passes through the active clamp switch and controls the active clamp switch based on the received active clamp switch current. The active clamp controller circuit is configured to enable the active clamp switch based on a first amplitude comparison, the first amplitude comparison being based on the active clamp switch current. The active clamp controller circuit is configured to disable the active clamp switch based on a second amplitude comparison and a third amplitude comparison, the second amplitude comparison and the third amplitude comparison being based on the active clamp switch current.

In some embodiments, an apparatus includes a power converter circuit to convert an input voltage from a voltage source to an output voltage. The power converter circuit includes a transformer which has a primary winding and a secondary winding, a first winding node of the primary winding being coupled to the voltage source to receive the input voltage. The apparatus includes a main switch that is coupled to a second winding node of the primary winding to control a current through the primary winding, and an active clamp circuit to clamp a voltage at the second winding node of the primary winding. The active clamp circuit includes: i) a series circuit combination of an active clamp capacitor coupled in series to an active clamp switch, the series circuit combination being coupled in parallel with the primary winding, and ii) an active clamp controller circuit to receive an active clamp switch current that passes through the active clamp switch and to control the active clamp switch based on the active clamp switch current. The active clamp controller circuit is configured to enable the active clamp switch based on a first amplitude comparison, the first amplitude comparison being based on the active clamp switch current. The active clamp controller circuit is configured to disable the active clamp switch based on a second amplitude comparison and a third amplitude comparison, the second amplitude comparison and the third amplitude comparison being based on the active clamp switch current.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a simplified circuit schematic of a conventional power converter.

FIG. 2 is a simplified circuit schematic of a power converter with a self-driven active clamp circuit, in accordance with some embodiments.

FIG. 3 is a simplified circuit schematic of a self-driven active clamp circuit, in accordance with some embodiments.

FIG. 4 is a simplified plot of signals related to a self-driven active clamp circuit, in accordance with some embodiments.

FIG. 5A is a simplified plot of signals related to a power converter with a self-driven active clamp circuit, in accordance with some embodiments.

FIG. 5B is a simplified plot of signals related to a power converter with a conventional clamp circuit.

FIG. 6 is a simplified circuit schematic of a portion of a power converter with a self-driven active clamp circuit, in accordance with some embodiments.

FIG. 7 is a simplified circuit schematic of a portion of a power converter with a self-driven active clamp circuit, in accordance with some embodiments.

FIG. 8 is a simplified circuit schematic of a portion of a power converter with a self-driven active clamp circuit, in accordance with some embodiments.

FIG. 9A is a simplified plot of signals related to a power converter with a self-driven active clamp circuit, in accordance with some embodiments.

FIG. 9B is a simplified plot of signals related to a power converter with a conventional clamp circuit.

FIG. 10 is a screenshot showing experimental results of a power converter with a self-driven active clamp circuit, in accordance with some embodiments.

FIG. 11 is a portion of a process for clamping a voltage of a main switch of a power converter using a self-driven active clamp circuit, in accordance with some embodiments.

FIG. 12 is a portion of a process for clamping a voltage of a main switch of a power converter using a self-driven active clamp circuit, in accordance with some embodiments.

 DETAILED DESCRIPTION

Some embodiments described herein provide a self-driven active clamp circuit and self-driven active clamping methods for use in a power converter that converts an input voltage to an output voltage using a transformer. In some embodiments, the self-driven active clamp described herein replaces a diode of a resistor-capacitor-diode (RCD) snubber circuit of a conventional power converter, advantageously converting the conventional power converter into a power converter having self-driven active clamping without needing to change additional control circuits of the power converter (such as a primary-side power management integrated circuit, or a secondary side synchronous switch controller integrated circuit). In other embodiments, the self-driven active clamp circuit is integrated into an initial design of a power converter.

The self-driven active clamp circuit advantageously increases power processing efficiency of a power converter by recycling energy stored in a leakage inductance of the transformer. To further improve light-to-medium power processing efficiency, in some embodiments, the active clamp circuit is advantageously disabled during light-load and low-line operation of the power converter (e.g., when a drain-source voltage of a primary side switch is below a voltage threshold, such as 500V).

In accordance with some embodiments, the self-driven active clamp circuit clamps a primary side peak voltage of a main switch, which enables the power converter to utilize primary side and/or secondary side switches having a lower voltage rating, leading to reduced power losses during switch conduction and/or switching. In some embodiments, the self-driven active clamp circuit turns on when a body-diode of an active clamp field effect transistor (FET) switch of the self-driven active clamp circuit begins conducting (e.g., a source-drain current of the switch is flowing through the switch) and turns off when near a second zero crossing of the source-drain current through the switch. Thus, an on-time of the self-driven active clamp circuit can be advantageously reduced as compared to conventional active clamp solutions because the active clamp switching does not need to coincide with a main switch turn-on time. As a result, a smaller active clamp capacitor can be utilized as compared to conventional clamping circuits because the active clamp circuit resonant period is shorter. Because the active clamp capacitor is a high voltage component, using a smaller capacitance can result in significant cost reduction benefits.

At light loads, zero-voltage-switching is conventionally difficult to achieve using conventional active clamp circuits because the period between main switch turn-on times is significantly longer. Embodiments of the self-driven active clamp circuit described herein allow for near zero-voltage-switching of the main switch on the primary side of the power converter across a wide range of output load currents, without the need for accurate modulation of the transformer magnetizing inductance current value. Such embodiments induce a low-amplitude resonant magnetizing inductance current, via discharging of the active clamp capacitor, which amplifies the quasi-resonant voltage amplitude. This advantageously results in near zero-voltage switching of the main-switch during quasi-resonant operation at the first valley, and significantly reduced main switch turn-on voltage at second and higher valleys.

A further advantage of the self-driven active clamp circuit described herein is a significantly lower active clamp RMS current (e.g., reduction of up to 10×), as compared to conventional active clamp zero-voltage-switching methods. This enables the self-driven active clamp circuit to utilize an active clamp switch having a higher RDSON (drain-source on-resistance) as compared to conventional active clamp circuits, resulting in an active clamp circuit which is lower cost, easier to integrate, and having a smaller physical size than conventional active clamp circuits.

FIG. 1 is a simplified circuit schematic of a conventional power converter (“converter”) 100. Some elements of the power converter 100 have been omitted from FIG. 1 to simplify the description of power converter 100 but are understood to be present. A voltage source Vin′ is received at the converter 100. Vin′ can be provided either as an alternating current (AC) or direct current (DC). An input side of the converter 100 generally includes an input voltage filter block 122, a rectifier block 116 (in the case of AC input), an input voltage buffer capacitor C1, an RCD snubber circuit block 114 (which includes a capacitor C2, a resistor R1 and a diode D1), a main switch M1, and a main switch controller circuit (“controller”) 118. The input voltage filter block 122, rectifier block 116 and the input buffer capacitor C1 provide a filtered, buffered, rectified, or otherwise conditioned input voltage Vin to a transformer 102.

The transformer 102 transfers power from the input side of the converter 100 to an output side of the converter 100 and generally includes a primary winding 104 with a first node 108 and a second node 110, and a secondary winding 106. The output side of the converter 100 generally includes an output buffer circuit 112, a synchronous rectifier switch M2, a synchronous rectifier switch controller circuit (“controller”) 120, and a load (not shown).

The first node 108 receives Vin, and the second node 110 is coupled to a terminal of the main switch M1 at the second node 110. The main switch M1 controls a current through the primary winding 104 to charge a magnetizing inductance LM of the transformer 102 during a first portion of a switching cycle of the converter 100. The synchronous rectifier switch M2 controls a current flow through the secondary winding 106 to discharge the transformer 102 into output buffer circuit 112 and/or a load during a subsequent portion of the switching cycle. In some embodiments, one or both of the main switch M1 and/or the synchronous rectifier switch M2 are field-effect transistors (FETs), each having a drain node, a source node, and a gate node to control a conduction of current between the drain node and the source node. In other embodiments, the synchronous rectifier switch M2 is replaced with a diode.

When the main switch M1 is enabled by the controller 118 during the first portion of a switching cycle, current flows through the primary winding 104 to a voltage bias node such as ground. The current flow through the primary winding 104 causes energy to be stored in the magnetization inductance LM and a leakage inductance LL of the transformer 102. When the main switch M1 is disabled in a subsequent portion of the switching cycle, output voltage Vout is generated at the output buffer circuit 112 and is provided to a load (not shown). When the main switch M1 is turned off, a reflected voltage (nVout) is developed at a drain node of the main switch M1 at node 110. The contribution of the reflected voltage nVout to a drain-source voltage Vds of the main switch M1 at the node 110 is expressed as:

V d ⁢ s = V in + nV o ⁢ u ⁢ t ( Equation ⁢ ⁢ 1 )
where n is a turns ratio of the transformer 102. Energy stored in the leakage inductance LL of the transformer 102 also contributes to the voltage Vds developed at the node 110.

The RCD snubber circuit 114 prevents the voltage Vds from increasing to a level that damages the main switch M1. As Vds rises, the diode D1 becomes forward biased and current flows into the capacitor C2 and into the resistor R1 to dissipate energy, thereby clamping Vds to a level that is within a safe operating range of the main switch M1. To further increase the efficiency of the converter 100, the diode D1 can be replaced with an actively driven clamp switch driven by an active clamp drive signal. However, conventional active clamping circuits require a control signal or other means of synchronization from the controller 118. Thus, a converter 100 that uses a controller 118 that is not already configured to support conventional active clamping cannot easily be modified to implement active clamping.

FIG. 2 is a simplified circuit schematic of a power converter (“converter”) 200 with a self-driven active clamp circuit 214, in accordance with some embodiments. Some elements of the power converter 200 have been omitted from FIG. 2 to simplify the description of the power converter 200 but are understood to be present. The converter 200 generally includes the circuit elements discussed with reference to FIG. 1. However, all, or a portion (e.g., the diode D1), of the snubber circuit 114 of the converter 100 has been replaced with the self-driven active clamp circuit (“active clamp circuit”) 214. Advantageously, the active clamp circuit 214 can replace the snubber circuit 114 of the converter 100 without making significant modifications to the converter 100 (e.g., it does not require a control signal or other synchronization signal from the controllers 118, 120). Thus, a converter 100 that was manufactured with a snubber circuit similar to the snubber circuit 114 can be modified with the self-driven active clamp circuit 214 to perform active clamping. For example, in some embodiments, the diode D1 of the conventional snubber circuit 114 can be replaced with the self-driven active clamp circuit 214. A magnetizing inductance current iLM flowing out of the primary winding 104 is described later with reference to FIG. 5A.

FIG. 3 is a simplified circuit schematic of the self-driven active clamp circuit 214 of the converter 200 introduced with reference to FIG. 2, in accordance with some embodiments. Some elements of the self-driven active clamp circuit 214 have been omitted from FIG. 3 to simplify the description of the self-driven active clamp circuit 214 but are understood to be present. The active clamp circuit 214 generally includes an active clamp switch controller circuit 302 which provides an active clamp switch control signal to the active clamp switch M3 via a gate driver circuit 314. The active clamp circuit 214 additionally includes a first current amplitude comparison circuit 304, a second current amplitude comparison circuit 306, a voltage peak detection and enable circuit 308, a voltage source circuit 316 (e.g., an LDO), an active clamp switch M3 having a body-diode D2, and an active clamp capacitor C3. The active clamp capacitor C3 is coupled in a series with the active clamp switch M3, and the series combination of the active clamp capacitor C3 and the active clamp switch M3 are coupled in parallel with the primary winding 104. The first current amplitude comparison circuit 304, the second current amplitude comparison circuit 306, the voltage peak detection and enable circuit 308, and the active clamp switch controller circuit 302 are referred to herein as being of an active clamp controller circuit. The active clamp capacitor C3 can advantageously be of a lower voltage rating as compared to the voltage rating of capacitor C2 of a conventional RCD circuit 114, thus providing cost savings. For example, in an example embodiment, the active clamp capacitor C3 can have a 250V rating as compared to a 630V rating of the capacitor C2. In some embodiments, the resistor R1 of the conventional RCD 114 can advantageously be omitted from the active clamp circuit 214, providing further cost savings.

The voltage source circuit 316 receives a voltage 318 and increases, decreases, or otherwise conditions the voltage 318 to power the gate driver circuit 314 in order to drive (i.e., enable and disable) the active clamp switch M3. In some embodiments, the active clamp switch M3 is a current-bidirectional two-quadrant switch. In some embodiments, the active clamp switch M3 is a field-effect transistor (FET) having a drain node (i.e., a first switch node), a source node (i.e., a second switch node), and a gate node (i.e., a switch control node) to control a conduction of current between the drain node and the source node. The drain node and the source node of the active clamp switch M3 are in a series circuit combination with the active clamp capacitor C3. The gate node of the active clamp switch M3 controls a current between the drain node and the source node of the active clamp switch M3. In some embodiments, the active clamp switch M3 includes a diode, other than a body-diode, which is configured to pass a current between the source node and the drain node (in a first current direction) when the diode is forward biased (e.g., when sufficient voltage is developed across the source and drain of the active clamp switch M3). When the gate node of the active clamp switch M3 is driven by the switch control signal, the active clamp switch M3 passes current bidirectionally (e.g., in the first current direction, and/or a second current direction). In the first current direction, current flows from the primary winding 104, through the active clamp switch M3, and into the active clamp capacitor C3. In the second current direction, current flows from the active clamp capacitor C3, through the active clamp switch M3, and into the primary winding 104.

The voltage peak detection and enable circuit 308 generally includes a voltage comparison circuit 310, a diode D3 and a capacitor C4. The capacitor C4 couples a first input of the voltage comparison circuit 310 to the node 110 to receive the drain-source voltage Vds of the main switch M3. The voltage comparison circuit 310 receives a voltage threshold 312 (e.g., from a voltage source circuit, not shown) at a second input. In some embodiments, the voltage threshold 312 is a voltage level above which the main switch M3 could be damaged. In other embodiments, the voltage threshold 312 is a voltage level that corresponds to a heavy load operation of the converter 200. In still other embodiments, the voltage threshold is a voltage level that corresponds to medium or heavy load operation of the converter 200. The voltage peak detection and enable circuit 308 transmits an enable signal to the active clamp switch controller circuit 302 when a voltage across the main switch M3 is equal to or surpasses the voltage threshold 312. The voltage peak detection and enable circuit 308 transmits a disable signal to the active clamp switch controller circuit 302 when a voltage across the switch M3 does not equal or surpass the voltage threshold 312. Thus, the active clamp switch controller circuit is configured to disable the active clamp switch M3 in response to a voltage amplitude of a voltage developed across the main switch M1 being less than the voltage threshold 312 (e.g., 500 V), irrespective of a current surpassing the first current threshold 320. That is, for as long as the active clamp switch controller circuit 302 receives a de-asserted signal at the ENABLE input, the active clamp switch M3 is disabled and the output PWM signal is not transitioned based on signals received by the active clamp switch controller circuit 302 at the SET and RESET inputs from the current amplitude comparison circuits 304, 306. In contrast, when the voltage comparison signal indicates that the received voltage surpasses the voltage threshold 312 (ENABLE is asserted), the output PWM signal is based on the SET and RESET signals received, by the active clamp switch controller circuit 302, from the current amplitude comparison circuits 304, 306. Thus, in some embodiments, the self-driven active clamp circuit 214 is advantageously disabled during light load operation of the converter 200, further increasing power efficiency.

During a portion of the switching cycle when the main switch M1 and active clamp switch M3 are both off, an active clamp switch current isd flows from the primary winding 104, through the body-diode D2, to the active clamp capacitor C3. During a subsequent portion of the switching cycle when the main switch M1 is off and the active clamp switch M3 is on, the current isd oscillates between the active clamp capacitor C3, the magnetizing inductance LM, and other intended or parasitic reactive elements of the converter 200.

At a high level, the active clamp switch M3 is enabled if the current iSD surpasses a first current threshold 320 due to the current iSD flowing from the primary winding 104 to the active clamp capacitor C3 through the body diode D2 of the active clamp switch M3 (e.g., body diode conduction of the D2 is detected). The active clamp switch M3 is disabled if a second zero crossing of the current iSD is detected. A first zero-crossing of the current iSD is determined to have occurred if the current iSD is less than a second current threshold 322, due to the current iSD flowing from the active clamp capacitor C3 into the primary winding 104 through the enabled active clamp switch M3. The second zero-crossing of the current iSD is determined to have occurred if the current iSD is greater than the second current threshold 322 at a later time, due to current flowing from the primary winding 104 to the active clamp capacitor C3 through the enabled active clamp switch M3. Upon determining that the second zero-crossing of the current iSD has occurred, the active clamp switch M3 is disabled.

To elaborate, the first current amplitude comparison circuit 304 receives the current isd at a first input, receives or generates the first current threshold 320 at a second input (e.g., from a first current source circuit, not shown), compares the received current isd to the first current threshold 320, and transmits a first comparison signal to a SET input of the active clamp switch controller circuit 302. In some embodiments, the first current threshold 320 is a current amplitude that identifies body-diode conduction of the active clamp switch M3. In some embodiments, the first current threshold 320 is about 50 mA to 250 mA, such as 100 mA. The second current amplitude comparison circuit 306 receives the current isd at a first input, receives or generates the second current threshold 322 at a second input (e.g., from a second current source circuit, not shown), compares the received current isd to the second current threshold 322, and transmits second and third comparison signals to a RESET input of the active clamp switch controller circuit 302. In some embodiments, the second current threshold 322 is a current amplitude that identifies a first and second zero-crossing of the current isd, the RESET input of the active clamp switch controller circuit 302 receiving an asserted second comparison signal in response to the second current amplitude circuit 306 detecting the first zero-crossing of the current isd, and receiving a de-asserted third comparison signal in response to the second current amplitude circuit 306 detecting the second zero-crossing of the current isd. In some embodiments, the second current threshold 322 is about −300 mA to 0 mA, such as −50 mA. In some embodiments, the first zero-crossing of the current isd is determined by the current isd transitioning across the second current threshold 322 with a negative going slope, and the second zero-crossing of the current isd is determined by the current isd transitioning across the second current threshold 322 with a positive going slope.

Thus, a PWM signal to drive the active clamp switch M3 via the gate driver circuit 314 is generated by the active clamp switch controller circuit 302 based on the comparison result signals received at SET and RESET inputs from the comparison circuits 304, 306. The PWM signal is transmitted to the gate driver circuit 314 to control the active clamp switch M3. FIG. 4 provides an example plot 400 illustrating a relationship between the SET and RESET comparison signals 402, 406 and a resultant PWM signal output 404 of the active clamp switch controller circuit 302, in accordance with some embodiments.

As shown, a rising edge of the SET comparison signal 402 received from the first current amplitude comparison circuit 304 (due to detecting body-diode conduction of M3) triggers a first edge 404a of the PWM signal 404. A falling edge of the SET comparison signal 402 does not cause the PWM signal 404 to change state. Likewise, a rising edge of the RESET comparison signal 406 received from the second current amplitude comparison circuit 306 (due to detecting a first zero-crossing of the current iSD) does not cause the PWM signal 404 to change state. However, a falling edge of the RESET comparison signal 406 (due to detecting a second zero-crossing of the current iSD) triggers a second edge 404b of the PWM signal 404. Thus, in the embodiment shown, the active clamp switch M3 is enabled when the current isd surpasses the first current threshold 320 (e.g., when the current isd is flowing through the body diode D2 to the active clamp capacitor C3). The active clamp switch M3 is disabled when a second zero-crossing of the current iSD is detected (e.g., when the current isd is resonating due to the magnetizing inductance LM, the active clamp capacitor C3, and other reactive components).

In some embodiments, the converter 200 relies on an indirect resonant mechanism to achieve near zero-voltage turn-switching of the main switch M1. FIG. 5A shows a simplified graph 502 which provides operation details of the converter 200, in accordance with some embodiments. In comparison, FIG. 5B shows a simplified graph 550, which provides operation details of a converter having a conventional active clamp circuit. In both graph 502 and graph 550, the voltage Vds at the node 110 and a current flow iLM of the transformer 102 over time t is shown. The current iLM is shown in graph 502 as having an opposite polarity to that of the current isd introduced with reference to FIG. 3. As was described with reference to FIG. 3, the active clamp switch M3 enable signal (PWM) 506 is generated in response to an amplitude of the current flow iSD (negative iLM). The simplified graph 502 corresponding to the converter 200 illustrates advantageous zero-voltage switching of the active clamp switch M3 at region 504 when the body-diode D2 is conducting, zero-current switching of the active clamp switch M3 at region 508, and near-zero voltage switching of the main switch M1 at region 514. The near-zero voltage switching of the main switch M1 in region 514 is achieved via a second-order effect arising from the operation of the active clamp circuit 214 as described herein. Specifically, when the active clamp circuit 214 is operational, the active clamp capacitor C3 voltage is maintained at a voltage value approximately equal to the reflected output voltage. As a result, at the instant when the converter 200 enters region 514 (quasi-resonant mode) the voltage across the active clamp switch M3 is approximately zero and the equivalent circuit becomes an LC circuit as shown in the simplified circuit schematic 700 of FIG. 7 (L is due to the transformer 102 magnetizing inductance and C is the equivalent capacitance Ceq seen at node 110). Of note is that the equivalent capacitance Ceq as seen at node 110 is dominated by the respective capacitance of the active clamp switch CossM3 and the active clamp capacitor C3 capacitance (as shown in the simplified schematic 700 of FIG. 7), which are in the nF range (several orders of magnitude greater than that of a conventional converter 100 with the RCD snubber circuit block 114). Due to the relatively large equivalent capacitance Ceq, the resonant magnetizing inductance current amplitude, iLM1 (as shown in FIG. 7), is forced to a larger value (as shown in FIG. 9A) as compared to the case where a conventional RCD snubber is utilized and the voltage at the clamp capacitor is significantly larger than the reflected output voltage (as shown in FIG. 9B). The peak magnetizing inductance current can be approximated according to the well-known LC resonant conservation of energy equation,
Lm·iLM12=Ceq·vds2  (Equation 2),
where Ceq is the equivalent capacitance at node 110 and iLM1 is a magnetizing inductance current (shown in FIG. 7 as iLM1). Because the amplitude of the magnetizing inductance current is forced to a higher value initially, and because the equivalent capacitance Ceq as seen at node 110 falls to pF range once the main switch M1 drain-source voltage begins to fall (a simplified schematic 800 of an equivalent circuit at this point is shown in FIG. 8), the voltage value at the valleys is forced to a lower voltage value (e.g., 2-3× time less as compared to a conventional RCD circuit).

Furthermore, a width of the PWM pulse 506 is significantly and advantageously shorter than a PWM pulse 556 corresponding to a conventional active clamp with zero-voltage switching of the main switch M1 at region 564. An RMS current irms of the graph 502 is significantly less than an RMS current irms of the graph 550, due to the peak currents being of the same amplitude ipk but the resonant period (and duty-ratio D, of the on-time/switching period) of the graph 502 being an order of magnitude less than graph 550, as given by the equation,

i r ⁢ m ⁢ s 2 = i p ⁢ k 2 · D 3 , ( Equation ⁢ ⁢ 3 )
thus, a smaller active clamp capacitor C3 (on the order of nanofarads) can be used, as compared to a conventional active clamp circuit which may require an active clamp capacitor being on the order of hundreds of nanofarads. Based on the near zero voltage crossing at region 514, the main switch M1 is switched on (e.g., resulting in pulse M1EN 516) when voltage Vds at node 110 is equal to Vin−knVout, where k is an integer value.

FIGS. 6-8 highlight parasitic capacitive and inductive elements of the converter 200 which resonate during quasi-resonant operation of the converter 200, a portion of the switching cycle when the main switch M1 and the synchronous rectifier switch M2 are both off. A simplified circuit schematic 600 of FIG. 6 includes elements of the converter 200 introduced in FIG. 2, as well as a representation of magnetizing inductance LM 130, parasitic capacitance CossM3 of the active clamp switch M3, parasitic capacitance CossM1 of the main switch M1, and parasitic capacitance CossM2 of the synchronous rectifier switch M2. Also shown is an active clamp switch voltage VAC which is a drain-source voltage of the active clamp switch M3. The dominant capacitive elements along with the range of their capacitance values for two different active clamp switch voltages are illustrated in the simplified circuit schematics 700, 800 of FIGS. 7-8, respectively.

The simplified circuit schematic 700 of FIG. 7 illustrates particular capacitive and inductive elements of the converter 200 during a portion of the switching cycle when the voltage across the capacitance CossM3 is approximately equal to zero volts, the voltage across the active clamp capacitor C3 is approximately equal to nVout, and the voltage Vds at node 110 is equal to Vin+nVout. At this time, the combined capacitance of the active clamp capacitor C3 and the parasitic capacitance CossM3 of the active clamp switch M3 is on the order of nanofarads. The parasitic capacitance CossM1 of the main switch M1 is on the order of picofarads, and the parasitic capacitance CossM2 of the synchronous rectifier switch M2 is on the order of picofarads.

The simplified schematic 800 of FIG. 8 illustrates particular capacitive and inductive elements of the converter 200 during a portion of the switching cycle when the voltage across the capacitance CossM3 is approximately equal to negative nVout, and the voltage across the active clamp capacitor C3 is approximately equal to nVout, resulting in the voltage Vds at node 110 being less than the input voltage Vin. During this portion of the switching cycle, the active clamp capacitor C3 voltage is significantly larger than the reflected output voltage nVout. The combined capacitance of the active clamp capacitor C3 and the parasitic capacitance CossM3 of the active clamp switch M3 is on the order of picofarads. The parasitic capacitance CossM1 of the main switch M1 is on the order of picofarads, and the parasitic capacitance CossM2 of the synchronous rectifier switch M2 is on the order of picofarads. A current iLM2 of the magnetizing inductance 130 is also shown.

The large difference in the capacitive values is important because the initial large effective capacitance leads to higher magnetizing inductance current (during LC resonance):

I p ⁢ k = V p ⁢ e ⁢ a ⁢ k × C e ⁢ q L M ( Equation ⁢ ⁢ 4 )
where Vpeak is equal to the reflected voltage nVout on the primary side of the transformer 102, LM is the magnetizing inductance 130, and Ceq is the equivalent capacitance of the intended and parasitic capacitances as seen at node 110 shown in FIGS. 6-8. The large difference in the capacitive values occurs due to a large active clamp FET output capacitance CossM3 of the active clamp switch M3 when its drain-source voltage is zero, which is only the case when the active clamp circuit 214 is enabled (e.g., when energy stored in the active clamp capacitor C3 is recycled during each switching cycle).

A simplified graph 900 of FIG. 9A illustrates voltage Vds at node 110 and a magnetizing inductance current iLM (iLM1-2) of the magnetizing inductance 130 over time t during a quasi-resonant operation of the converter 200, in accordance with some embodiments. A simplified graph 950 of FIG. 9B is a plot of voltage Vds and current iLM over time t during a quasi-resonant portion of the switching cycle for a conventional lossy RCD snubber circuit. The simplified graph 900 illustrates the non-linear effect of Equation 4 when the active clamp capacitor C3 voltage VAC is approximately equal to the reflected output voltage (e.g., VAC˜nVout), as compared to the simplified graph 950 which illustrates a conventional lossy snubber operation where the snubber capacitor voltage is significantly greater than the reflected output voltage (e.g., VAC>1.5×nVout).

As shown in the simplified graph 900, the active clamp circuit 214 allows for a higher magnetizing current iLM as compared to the conventional lossy snubber circuit shown in the simplified graph 950. The higher magnetizing current iLM, in turn, enables more charge to be removed from the main switch M1, thus leading to lower drain-source voltages Vds during quasi-resonant operation of the converter 200.

FIG. 10 is a screenshot 1000 of experimental results of Vds over time for a converter 200 having an active clamp circuit 214, in accordance with some embodiments. A first portion 1002 of the screenshot 1000 shows voltage Vds at node 110 during operation of the converter 200 when the active clamp circuit 214 is disabled. A second portion 1052 of the screenshot 1000 shows voltage Vds at node 110 during operation of the converter 200 when the active clamp circuit 214 is enabled and the active clamp capacitor voltage VAC is approximately equal to the reflected output voltage nVout. As shown, when the active clamp 214 is enabled, the main switch M1 drain-source voltage at the valleys is significantly smaller (e.g., 100 Vdc vs. 225 Vdc at the first valley; 125 Vdc vs. 240 Vdc at the second valley and etc.). The benefit of the reduced valley voltage is due to the reduced main switch CossM1 energy dissipated, which is a function of the voltage squared, when the main switch M1 is turned on.

FIG. 11 is a portion of an example process for clamping a voltage of a main switch of a power converter (e.g., the power converter 200), in accordance with some embodiments. The particular steps, order of steps, and combination of steps are shown for illustrative and explanatory purposes only. Other embodiments can implement different particular steps, orders of steps, and combinations of steps to achieve similar functions or results. At step 1102, an input voltage is received at a primary winding (104) of a transformer (102) of the power converter (200). At step 1104, a current through the primary winding (104) is controlled using a main switch (M1) of the power converter (200). At step 1106, a voltage of the main switch (e.g., at node 110) is clamped to a maximum voltage using an active clamp circuit (214). The voltage of the main switch (M1) is clamped based on an active clamp switch current (iSD) that passes through an active clamp switch (M3) of the active clamp circuit (214). Details of step 1106 are presented in FIG. 12, in accordance with some embodiments. The particular steps, order of steps, and combination of steps are shown for illustrative and explanatory purposes only. Other embodiments can implement different particular steps, orders of steps, and combinations of steps to achieve similar functions or results.

At step 1202, the active clamp switch current (iSD) is received at a first current amplitude comparison circuit (304) at a first time. At step 1204, it is determined, using the first current amplitude comparison circuit, if the active clamp switch current is greater than a first current threshold (320) (e.g., 100 mA). If it is determined at step 1204 that the active clamp switch current is not greater than the first current threshold, flow returns to step 1202. If it is determined at step 1204 that the active clamp switch current is greater than the first current threshold (e.g., a body-diode of the active clamp switch M3 is conducting), flow continues to step 1206. At step 1206, the active clamp switch (M3) is enabled (e.g., turned on), thereby clamping a voltage at the main switch (M1). At step 1208, the active clamp switch current is received at a second current amplitude comparison circuit (306) at a second time. At step 1210, it is determined, using the second current amplitude comparison circuit, if the active clamp switch current is less than a second current threshold (322) (e.g., −50 mA). If it is determined at step 1210 that the active clamp switch current is not less than the second current threshold, flow returns to step 1208. If it is determined at step 1210 that the active clamp switch current is less than the second current threshold (e.g., a first zero crossing of the active clamp switch current), flow continues to step 1212. At step 1212, the active clamp switch current is received at the second current amplitude comparison circuit at a third time. At step 1214, it is determined, using the second current amplitude comparison circuit, if the active clamp switch current is greater than the second current amplitude comparison circuit. If it is determined at step 1214 that the active clamp switch current is not greater than the second current threshold, flow returns to step 1212. If it is determined at step 1214 that the active clamp switch current is greater than the second current threshold (e.g., a second zero crossing of the active clamp switch current), flow continues to step 1216. At step 1216, the active clamp switch (M3) is disabled (e.g., it is turned off). When the active clamp switch is turned off, the active clamp circuit (214) is no longer clamping a voltage of a main switch (M1) of the power converter (200).

Reference has been made in detail to embodiments of the disclosed invention, one or more examples of which have been illustrated in the accompanying figures. Each example has been provided by way of explanation of the present technology, not as a limitation of the present technology. In fact, while the specification has been described in detail with respect to specific embodiments of the invention, it will be appreciated that those skilled in the art, upon attaining an understanding of the foregoing, may readily conceive of alterations to, variations of, and equivalents to these embodiments. For instance, features illustrated or described as part of one embodiment may be used with another embodiment to yield a still further embodiment. Thus, it is intended that the present subject matter covers all such modifications and variations within the scope of the appended claims and their equivalents. These and other modifications and variations to the present invention may be practiced by those of ordinary skill in the art, without departing from the scope of the present invention, which is more particularly set forth in the appended claims. Furthermore, those of ordinary skill in the art will appreciate that the foregoing description is by way of example only and is not intended to limit the invention.

Claims

1. A method, comprising:

receiving, at an active clamp controller circuit, an active clamp switch current that passes through an active clamp switch;
enabling, using the active clamp controller circuit, the active clamp switch in response to determining, based on the active clamp switch current, body-diode conduction of the active clamp switch; and
disabling, using the active clamp controller circuit, the active clamp switch in response to determining, based on the active clamp switch current, a first zero-crossing of the active clamp switch current and a second zero-crossing of the active clamp switch current.

2. The method of claim 1, wherein determining body-diode conduction of the active clamp switch comprises:

comparing, at a first amplitude comparison circuit of the active clamp controller circuit, the active clamp switch current to a first current amplitude threshold; and
generating, using the first amplitude comparison circuit, a first comparison signal based on the comparison of the active clamp switch current to the first current amplitude threshold, a level of the first comparison signal being indicative of body-diode conduction of the active clamp switch, the active clamp switch being enabled based on the first comparison signal.

3. The method of claim 2, wherein determining the first zero-crossing of the active clamp switch current and the second zero-crossing of the active clamp switch current comprises:

comparing, at a second amplitude comparison circuit of the active clamp controller circuit, the active clamp switch current to a second current amplitude threshold; and
generating, using the second amplitude comparison circuit, a second comparison signal based on the comparison of the active clamp switch current to the second current amplitude threshold, a first level of the second comparison signal being indicative of the first zero-crossing of the active clamp switch and a second level of the second comparison signal being indicative of the second zero-crossing of the active clamp switch, the active clamp switch being disabled based on the second comparison signal.

4. The method of claim 3, wherein:

determining the first zero-crossing of the active clamp switch current comprises comparing the active clamp switch current to the second current amplitude threshold at a first time; and
determining the second zero-crossing of the active clamp switch current comprises comparing the active clamp switch current to the second current amplitude threshold at a second time.

5. The method of claim 1, further comprising:

clamping, by the active clamp switch, a voltage at a node of another switch to a maximum voltage, the active clamp switch being coupled to the node of the other switch.

6. The method of claim 5, wherein:

the node of the other switch is coupled to a winding of a transformer; and
the active clamp switch current is a current that flows bi-directionally, through the active clamp switch, between the winding of the transformer and an active clamp capacitor.

7. The method of claim 1, further comprising:

determining, by the active clamp controller circuit, that a voltage amplitude of a switch voltage developed at a node of another switch is less than a voltage threshold; and
disabling, by the active clamp controller circuit, the active clamp switch in response to determining that the voltage amplitude of the switch voltage is less than the voltage threshold, irrespective of determining body-diode conduction of the active clamp switch.

8. The method of claim 7, further comprising:

receiving, at a voltage comparison circuit, the switch voltage;
generating, using the voltage comparison circuit, a voltage comparison signal based on a comparison between the voltage amplitude of the switch voltage and the voltage threshold;
disabling, by the active clamp controller circuit, the active clamp switch in response to the voltage comparison signal when the voltage comparison signal indicates that the voltage amplitude of the switch voltage does not surpass the voltage threshold; and
not disabling, by the active clamp controller circuit, the active clamp switch in response to the voltage comparison signal when the voltage comparison signal indicates that the voltage amplitude of the switch voltage surpasses the voltage threshold.

9. A method, comprising:

receiving, at a power converter circuit, an input voltage from a voltage source;
converting, by the power converter circuit, the input voltage to an output voltage, the power converter circuit comprising a transformer having a primary winding and a secondary winding, a first winding node of the primary winding being coupled to the voltage source to receive the input voltage;
controlling, by a main switch of the power converter circuit, a current through the primary winding, the main switch being coupled to a second winding node of the primary winding;
receiving, at an active clamp controller circuit, an active clamp switch current that passes through an active clamp switch that is coupled to a node of the main switch;
enabling, using the active clamp controller circuit, the active clamp switch in response to determining, based on the active clamp switch current, body-diode conduction of the active clamp switch; and
disabling, using the active clamp controller circuit, the active clamp switch in response to determining, based on the active clamp switch current, a first zero-crossing of the active clamp switch current and a second zero-crossing of the active clamp switch current.

10. The method of claim 9, wherein determining body-diode conduction of the active clamp switch comprises:

comparing, at a first amplitude comparison circuit of the active clamp controller circuit, the active clamp switch current to a first current amplitude threshold; and
generating, using the first amplitude comparison circuit, a first comparison signal based on the comparison of the active clamp switch current to the first current amplitude threshold, a level of the first comparison signal being indicative of body-diode conduction of the active clamp switch, the active clamp switch being enabled based on the first comparison signal.

11. The method of claim 10, wherein determining the first zero-crossing of the active clamp switch current and the second zero-crossing of the active clamp switch current comprises:

comparing, at a second amplitude comparison circuit of the active clamp controller circuit, the active clamp switch current to a second current amplitude threshold; and
generating, using the second amplitude comparison circuit, a second comparison signal based on the comparison of the active clamp switch current to the second current amplitude threshold, a first level of the second comparison signal being indicative of the first zero-crossing of the active clamp switch, a second level of the second comparison signal being indicative of the second zero-crossing of the active clamp switch, the active clamp switch being disabled based on the second comparison signal.

12. The method of claim 11, wherein:

determining the first zero-crossing of the active clamp switch current comprises comparing the active clamp switch current to the second current amplitude threshold at a first time; and
determining the second zero-crossing of the active clamp switch current comprises comparing the active clamp switch current to the second current amplitude threshold at a second time.

13. The method of claim 9, further comprising:

clamping, by the active clamp switch, a voltage at the node of the main switch to a maximum voltage.

14. The method of claim 9, wherein:

the active clamp switch current is a current that flows bi-directionally through the active clamp switch between the primary winding of the transformer and an active clamp capacitor.

15. The method of claim 9, further comprising:

disabling, using the active clamp controller circuit, the active clamp switch in response to determining that a voltage amplitude of a switch voltage developed at the node of the main switch is less than a voltage threshold, irrespective of determining body-diode conduction of the active clamp switch.

16. The method of claim 15, further comprising:

receiving, by a voltage comparison circuit, the switch voltage;
outputting, by the voltage comparison circuit, a voltage comparison signal based on a comparison between the voltage amplitude of the switch voltage and the voltage threshold;
disabling the active clamp switch in response to the voltage comparison signal when the voltage comparison signal indicates that the voltage amplitude of the switch voltage does not surpass the voltage threshold; and
not disabling the active clamp switch in response to the voltage comparison signal when the voltage comparison signal indicates that the voltage amplitude of the switch voltage surpasses the voltage threshold.

17. An apparatus, comprising:

an active clamp switch; and
an active clamp controller circuit coupled to the active clamp switch and being operable to receive an active clamp switch current that passes through the active clamp switch;
wherein:
the active clamp controller circuit is operable to enable the active clamp switch in response to determining, based on the active clamp switch current, body-diode conduction of the active clamp switch; and
the active clamp controller circuit is operable to disable the active clamp switch in response to determining, based on the active clamp switch current, a first zero-crossing of the active clamp switch current and a second zero-crossing of the active clamp switch current.

18. The apparatus of claim 17, further comprising:

a first amplitude comparison circuit of the active clamp controller circuit;
wherein:
the first amplitude comparison circuit is operable to compare the active clamp switch current to a first current amplitude threshold; and
the active clamp controller circuit is operable to generate a first comparison signal based on the comparison of the active clamp switch current to the first current amplitude threshold, a level of the first comparison signal being indicative of body-diode conduction of the active clamp switch, the active clamp switch being enabled based on the first comparison signal.

19. The apparatus of claim 18, further comprising:

a second amplitude comparison circuit of the active clamp controller circuit;
wherein:
the second amplitude comparison circuit is operable to compare the active clamp switch current to a second current amplitude threshold; and
the active clamp controller circuit is operable to generate a second comparison signal based on the comparison of the active clamp switch current to the second current amplitude threshold, a first level of the second comparison signal being indicative of the first zero-crossing of the active clamp switch, a second level of the second comparison signal being indicative of the second zero-crossing of the active clamp switch, the active clamp switch being disabled based on the second comparison signal.

20. The apparatus of claim 17, further comprising:

a third amplitude comparison circuit of the active clamp controller circuit;
wherein:
the third amplitude comparison circuit is operable to compare a voltage amplitude of a switch voltage at a node of another switch to a voltage threshold; and
the active clamp controller circuit is operable to disable the active clamp switch in response determining that the voltage amplitude of the switch voltage is less than the voltage threshold, irrespective of determining body-diode conduction of the active clamp switch.
Referenced Cited
U.S. Patent Documents
5455757 October 3, 1995 Nguyen et al.
5796595 August 18, 1998 Cross
6069803 May 30, 2000 Cross
6452818 September 17, 2002 Simopoulos
6507174 January 14, 2003 Qian
RE38196 July 22, 2003 Vinciarelli et al.
7573731 August 11, 2009 Kwon
7606051 October 20, 2009 Wittenbreder
8358518 January 22, 2013 Yang et al.
9246391 January 26, 2016 Berghegger
10673342 June 2, 2020 Tzeng
10965218 March 30, 2021 Odell
20030179592 September 25, 2003 Nishiyama et al.
20050201128 September 15, 2005 Jin et al.
20050285661 December 29, 2005 Wittenbreder
20060013021 January 19, 2006 Aso
20060062026 March 23, 2006 Wittenbreder
20070195560 August 23, 2007 Yasumura
20070263415 November 15, 2007 Jansen et al.
20080123374 May 29, 2008 Vinciarelli
20080315858 December 25, 2008 Hong et al.
20090073725 March 19, 2009 Lin
20110194313 August 11, 2011 Yoshinaga
20110305048 December 15, 2011 Yang et al.
20140185333 July 3, 2014 Yang
20140233275 August 21, 2014 Yang et al.
20140268914 September 18, 2014 Wang et al.
20140307484 October 16, 2014 Yang
20150003121 January 1, 2015 Yang et al.
20150016153 January 15, 2015 Orr
20150103567 April 16, 2015 Wang et al.
20150263639 September 17, 2015 Russell et al.
20150318788 November 5, 2015 Kovacic et al.
20150357904 December 10, 2015 Odell et al.
20160072399 March 10, 2016 Kikuchi et al.
20160099647 April 7, 2016 Zhang et al.
20160226389 August 4, 2016 Quaglino et al.
20160344293 November 24, 2016 Hari
20160365801 December 15, 2016 Phadke
20170070152 March 9, 2017 Liu
20170288528 October 5, 2017 McCoy et al.
20180115252 April 26, 2018 Chang et al.
20180175725 June 21, 2018 Hu et al.
20180226895 August 9, 2018 Song et al.
20180254710 September 6, 2018 Wang et al.
20180287481 October 4, 2018 Liu et al.
20180294735 October 11, 2018 Song et al.
20180301999 October 18, 2018 Moon et al.
20190013739 January 10, 2019 Hari et al.
20190044449 February 7, 2019 Li et al.
20190044450 February 7, 2019 Muto et al.
20190052174 February 14, 2019 Gong
20190131963 May 2, 2019 Gong et al.
20190149052 May 16, 2019 Gong et al.
20190149054 May 16, 2019 Shimura et al.
20200007041 January 2, 2020 Liu et al.
20200112261 April 9, 2020 Strijker
20200227995 July 16, 2020 Radic
20200366214 November 19, 2020 Radic
20210152094 May 20, 2021 Wong
20210305889 September 30, 2021 Radic
Foreign Patent Documents
100525045 August 2009 CN
102208873 October 2011 CN
103795260 May 2014 CN
108683336 October 2018 CN
108683336 October 2019 CN
2009290932 December 2009 JP
2014143209 August 2014 JP
20160125676 November 2016 KR
20170106204 September 2017 KR
20170108869 September 2017 KR
201315118 April 2013 TW
2016171325 October 2016 WO
2017095408 June 2017 WO
Other references
  • Notice of Allowance and Fees dated Jun. 1, 2022 for U.S. Appl. No. 17/304,042.
  • International Search Report dated Apr. 26, 2019 for PCT Patent Application No. PCT/IB2018/060252.
  • International Search Report dated Jul. 27, 2020 for PCT Patent Application No. PCT/IB2020/053739.
  • International Search Report dated May 31, 2019 for PCT Patent Application No. PCT/IB2019/050686.
  • Notice of Allowance and Fees dated Jun. 17, 2021 for U.S. Appl. No. 16/889,194.
  • Notice of Allowance dated Apr. 17, 2019 for U.S. Appl. No. 16/221,945.
  • Notice of Allowance dated Apr. 23, 2021 for U.S. Appl. No. 16/913,489.
  • Notice of Allowance dated Apr. 3, 2020 for U.S. Appl. No. 16/569,924.
  • Notice of Allowance dated Feb. 26, 2021 for U.S. Appl. No. 16/665,237.
  • Notice of Allowance dated Feb. 7, 2020 for U.S. Appl. No. 16/401,980.
  • Notice of Allowance dated Jan. 5, 2021 for Taiwanese Patent application No. 107145962.
  • Notice of Allowance dated Jul. 3, 2019 for U.S. Appl. No. 16/247,227.
  • Notice of Allowance dated Mar. 6, 2020 for U.S. Appl. No. 16/145,819.
  • Office Action dated Jul. 29, 2019 for U.S. Appl. No. 16/145,819.
  • Office Action dated Jul. 30, 2020 for U.S. Appl. No. 16/665,237.
  • Office Action dated May 31, 2019 for U.S. Appl. No. 16/247,227.
  • Office Action dated Nov. 4, 2020 for U.S. Appl. No. 16/665,237.
  • Office Action dated Oct. 3, 2019 for U.S. Appl. No. 16/401,980.
  • Official Letter and Search Report dated Aug. 5, 2020 for Taiwan Patent application No. 107145962.
  • Office Action dated Sep. 2, 2021 for U.S. Appl. No. 16/944,270.
Patent History
Patent number: 11671026
Type: Grant
Filed: Aug 16, 2021
Date of Patent: Jun 6, 2023
Patent Publication Number: 20210376747
Assignee: Appulse Power Inc. (Toronto)
Inventor: Aleksandar Radic (Toronto)
Primary Examiner: Matthew V Nguyen
Application Number: 17/445,117
Classifications
Current U.S. Class: With Additional Series Regulator (323/224)
International Classification: H02M 3/335 (20060101); H02M 1/08 (20060101); H02M 1/34 (20070101);